Blind separation and deconvolution of MIMO-FIR system with colored inputs based on SIMO-model-based ICA

The paper describes a novel two-stage blind separation and deconvolution algorithm for a multiple-input multiple-output FIR system driven by colored sources, in which a new single-input multiple-output (SIMO)-model-based ICA (SIMO-ICA) and blind multichannel inverse filtering are combined. SIMO-ICA can separate the mixed signals, not into monaural source signals, but into SIMO-model-based signals from independent sources. After SIMO-ICA, all SIMO blind deconvolution can be applied, even when each source signal is temporally correlated. Experimental results reveal that the proposed algorithm can successfully achieve separation and deconvolution for a convolutive mixture of speech signals.